#90350: Neuromodulation for Refractory Neuropsychiatric and Pain Disorders

First demonstrated in 1985, TMS projects a fluctuating magnetic field (magnetic pulses) through the skull into the brain via an inductor coil placed against the scalp. This generates electrical currents in brain tissue (via electromagnetic induction) of sufficient strength to modulate neuronal firing. Multiple TMS pulses given consecutively are termed repetitive TMS. rTMS is noninvasive, and unlike ECT does not require anesthesia [2,8].

Frequency

Frequency is the intensity of stimulation energy [2]. High-frequency (5–20 Hz) rTMS (HF-rTMS) increases neuronal activity and cortical excitability and increases relative regional cerebral blood flow (with exceptions). Low-frequency (≤1 Hz) rTMS (LF-rTMS) reduces neuronal activity and cortical excitability. HF-rTMS and LF-rTMS, applied to the same brain site, can induce dissimilar effects on brain circuits.

Stimulation strength is tailored to the motor threshold of each patient that is determined by the amount of current, with single-pulse TMS applied to the primary motor cortex (M1), that produces a twitch in a peripheral muscle (a motor-evoked potential) [11,12]. rTMS is usually applied in "trains" of pulses interspersed with rest periods, because prolonged high-frequency stimulation increases seizure risk [13].

Coil Type

Most rTMS studies use a standard figure-8 coil, which has the drawback of poor brain penetration that limits the depth of stimulation delivery, although superficial areas of the prefrontal cortex (PFC) such as the dorsolateral PFC represent important targets fully amenable to rTMS stimulation. This partially explains the preponderance of rTMS studies in MDD, as the dorsolateral PFC is highly relevant in depression [12].

The deep (H) coil delivers pulses deeper than the figure-8 coil to stimulate broader and deeper frontal cortex regions such as the insula and ventrolateral prefrontal cortex, as well as the dorsolateral PFC. This may come at the expense of specificity, where deeper penetration expands the magnetic field to stimulate regions and circuits that interfere with therapeutic response or contribute to adverse effects [2,12].

Mechanism

Trains of magnetic pulses with rTMS temporarily summate to alter neural activity that can modulate cortical excitability. LF-rTMS mostly stimulates low-threshold inhibitory interneurons, and HF-rTMS excite projection neurons [13]. The effects of rTMS are not limited to the brain areas under the stimulation coil, and remote brain areas connected to the stimulated site are affected [14]. Because of synaptic connections, distal effects (cortical and subcortical, ipsilateral and contralateral) on neural activity, regional cerebral blood flow, and neurotransmitter activity may differ from proximal effects. rTMS can be viewed as targeting brain circuits instead of neurotransmitters, although behavioral effects depend on neurotransmitter systems within the manipulated brain circuit [2]. Long-lasting therapeutic effects with rTMS are related to long-term potentiation and long-term depression mediated through N-methyl-D-aspartate (NMDA) synaptic plasticity [15].

tDCS is a non-invasive, painless brain stimulation treatment in numerous psychiatric and chronic pain conditions. tDCS is delivered through a device that transfers low-intensity (0.5–3 mA) electrical current to the scalp surface with 2 large (20–35 cm²2) saline-soaked sponge-electrodes, an anode and a cathode. The anode and cathode are attached to distinct areas of the scalp with a rubber headband. The current penetrates the skull and enters the brain from the anode, travels through the tissue, and exits via the cathode [16].

In contrast to other non-invasive brain stimulation modes, tDCS does not induce neuron action potentials but instead modulates neuron membrane excitability [15]. Parameters that influence patient response and functional outcomes with tDCS are current polarity, delivered dose, and electrode positions.

Stimulation Types and Effects

The tDCS current is too weak to induce neuron firing, but alters neuron excitability by shifting the resting membrane potentials of neurons in a depolarizing or hyperpolarizing direction, making them more or less likely to re-fire. Changes in spontaneous neuron firing induced by stimulation type [4,10]:

Thus, effects of tDCS on cortical excitability are polarity-dependent (anodal or cathodal), and both cause changes in spontaneous firing.

Mechanisms

With repeated use, tDCS modulates synaptic connectivity by inducing long-term potentiation and depression, mediated in part by NMDA receptor-dependent mechanisms; neuroplastic changes regulated by neurotransmitters systems such as dopamine, acetylcholine, gamma-aminobutryic acid (GABA), serotonin and BDNF; and changes in neuronal membrane channels including sodium and calcium pumps [8,17,18]. tDCS can also upregulate and downregulate functional connectivity within brain networks, including those important for cognitive, motor, and pain processing [18].

Theta-burst stimulation is a form of rTMS that delivers a 50-Hz burst of three pulses every 200 ms [19]. Intermittent theta-burst stimulation delivers 600 pulses of stimulation for 2 seconds, followed by an 8-second rest, over two to three minutes. Intermittent theta-burst stimulation is considered excitatory. In contrast, continuous theta-burst stimulation is applied in 40 second trains to result in long-term depression-like decreases in motor cortex excitability. Continuous theta-burst stimulation is considered inhibitory [8].

Magnetic seizure therapy is a noninvasive convulsive neurostimulation therapy currently investigated as an alternative to ECT. A neurostimulator and coil are placed directly on the skull. Electrical current passing through the coil generates a strong focal magnetic field (2 Tesla). This passes unimpeded through the skull and soft tissue to reach brain tissue, where an electrical current is induced that causes neuronal depolarization triggering a generalized tonic-clonic seizure. Magnetic seizure therapy and ECT both induce seizures in patients under general anesthesia with assisted ventilation and EEG monitoring, but magnetic seizure therapy may have fewer side effects, particularly cognitive impairment [8,20].

An early approach to brain stimulation, transcranial electrical stimulation applied high-voltage electric stimuli through electrodes on the scalp. Most of the current travelled along the scalp between the electrodes, but a small portion penetrated the brain to activate neurons. This method was refined in 1980 and termed transcranial electrical stimulation. Transcranial electrical stimulation is credited as the pioneering neurophysiological approach for noninvasive brain stimulation, but interest in transcranial electrical stimulation declined rapidly with the introduction of TMS in 1985 [19].

Most published research on transcranial electrical stimulation efficacy has involved tDCS. Only recently has tACS attracted attention as a promising alternative approach to directly interact with ongoing oscillatory cortical activity. To date, evaluation of tACS has been primarily limited to possible cognitive enhancing effects in healthy adults, and in the elderly with normal, age-related cognitive declines [21].

tRNS is a variant of tACS that uses a constantly changing current. While tACS stimulates at a set frequency to entrain oscillatory activity, tRNS stimulates at a randomly changing frequency between 0.1–640 Hz. tRNS at higher frequencies (>100 Hz) may induce larger excitability changes than stimulation with a direct current such as tDCS, because the sodium channels of underlying neurons are repeatedly opened by stimulation. Neuronal homeostatic mechanisms are prevented because the underlying neurons cannot adjust to the constantly randomly changing electrical field. As a relatively new technique, the number of published reports on tRNS treatment in psychiatric and pain disorders is sparse [22].

eTNS is an external neuromodulatory device originally developed for use in refractory seizures. The theoretical basis for eTNS is that bilateral electrical stimulation over the trigeminal cranial nerve may neuromodulate cortical and subcortical areas related to neuropsychiatric disorders, such as the amygdala, insular cortex, prefrontal cortex, hippocampus, thalamus, locus coeruleus, nucleus of the solitary tract, and the anterior cingulate cortex. eTNS involves a bottom-up mechanism where stimuli propagate from the cranial nerve in the direction of the brainstem and central brain regions. As such, this represents a novel approach that differs from transcranial stimulation involving top-down mechanisms [23,24].

Two eTNS protocols are described for psychiatric disorder treatment. Both protocols use stimulation frequencies of 120 Hz, asymmetrical alternate current, pulse duration of 0.25 ms and intensity based on patient sensitivity and pain threshold. Protocol I delivers eTNS in 8-hour sessions during sleep for 8 weeks, with a 30 seconds on/30 seconds off cycle, while Protocol II uses 30-minute sessions in 10 weekdays for 2 weeks, with continuous stimulation [24].

The concept of using focused ultrasound beams deep in the brain to treat movement disorders was conceived in the 1950s, but the need to make a cranial window in the skull led to its development abandoned. Low-intensity focused ultrasound can be delivered through the intact skull, making ultrasound neuromodulation the focus of considerable interest [7]. An advantage is its ability for deep brain delivery without causing permanent damage or effects [25]. Potential indications include acute symptoms such as seizures, and chronic conditions such as depression where plasticity may be important. Ultrasound neuromodulation is at a much earlier stage of development than TMS or tDCS [7].

VNS is a minimally invasive therapy, first approved in 1997 for refractory seizure disorders and in 2005 for treatment-refractory MDD. Vagus nerves are cranial nerve with fibers that transmit nerve impulses from the periphery to the brain. Electrical stimulation of the vagus nerve stimulates the nucleus tractus solitaries, enhancing its capacity to modulate brain areas through its neuronal connections distributed to subcortical and cortical brain regions [8].

With VNS, an implantable pulse generator is surgically inserted in the chest region. The lead from the device is attached to the left vagus nerve in the neck, where a small electrical current is delivered. Stimulation is around-the-clock, with typical stimulation periods or "trains" lasting 30 seconds, with 5 minutes of rest between trains. The device parameters (pulse width, frequency, duration of stimulation, and duty cycle) can be modified transcutaneously (as with cardiac pacemakers) using a wand attached to a small handheld programming computer [26].

Motor cortex stimulation is an invasive procedure, with use in chronic refractory pain first documented in the early 1990s. With motor cortex stimulation, the skull is surgically opened and an electrode array is inserted [13]. In epidural motor cortex stimulation, the electrodes are attached to the dura directly above the motor cortex. In subdural motor cortex stimulation, the dural layer is penetrated for electrode placement.

Motor cortex stimulation affects structures involved in affective, cognitive, and emotional aspects of pain. The analgesic mechanisms of motor cortex stimulation involve extensive changes at various CNS levels, with a likely entry point at the thalamic level that triggers distant effects on limbic, cingulate and orbitofrontal cortices, and descending modulation that reaches the spinal horn. Activation of the endogenous opioid system or gamma-aminobutyric acid transmission in these areas is thought to underlie long-term pain relief with motor cortex stimulation [13,27].

Deep brain stimulation is an invasive neurosurgical procedure performed under MRI guidance, where electrodes are implanted through the skull and dura into discrete brain targets for stimulation. The electrodes are internalized and connected to an implantable pulse generator implanted into the chest below the right clavicle. The implantable pulse generator is accessible with a handheld device, which allows remote monitoring and/or programming of stimulation parameters. The deep brain stimulation parameters of pulse width, frequency, and amplitude (voltage or current) are programmed by the treating physician and titrated to clinical effect. Deep brain stimulation has been extensively developed and refined for use in later-stage Parkinson disease, where it is a standard of care. Deep brain stimulation is also used in other movement disorders, but severe refractory psychiatric and pain disorders represent a growing research field [8].

Generalized anxiety disorder (GAD) is a chronic psychiatric condition defined by excessive and uncontrollable worry, with at least three of six hyperarousal symptoms (restlessness, muscle tension, fatigue, irritability, difficulty in sleeping, concentration problems). Lifetime prevalence is 5.7%, with higher rates in primary care and psychiatric outpatient samples. GAD is frequently comorbid with depression, complicating presentation, treatment and prognosis. At the individual level, GAD is associated with significant quality-of-life impairments and diminished physical health. At the systems level, GAD is associated with high use of health care services and high costs [33].

Pharmacotherapy (antidepressants, anxiolytics) and cognitive-behavioral therapy are standard treatments for GAD, but fail to achieve symptom remission in 33% to 50% of patients. The best existing treatments leave many patients with GAD without relief, and alternative treatments are needed [33].

GAD is characterized by abnormalities in frontal and limbic structures primarily involving the prefrontal cortex, anterior cingulate cortex, and amygdala; decreased structural and functional connectivity between frontal and limbic regions; and impairments in emotion regulation associated with abnormal functioning in neural circuits that encompass frontolimbic regions including the dorsolateral PFC [34].

Panic disorder is characterized by sudden unexpected panic attacks and apprehension about the possible causes and consequences of the attacks. Comorbid agoraphobia is characterized by behavioral avoidance of situations from which escape might be difficult if a panic attack occurred [35].

Patients with panic disorder with and without agoraphobia have shown PFC hypoactivity paired with fear-relevant brain structure (amygdala) hyperactivity, suggesting inadequate PFC inhibition of responses to anxiety-related stimuli [36]. Hypofrontality is shown in patients with panic disorder during cognitive tasks without emotional content, and in responses to emotional stimuli. Altered activation patterns in the left inferior PFC are associated with panic attacks, and altered right anterior PFC associated with severity of agoraphobic avoidance [35,36,37].

A variety of scales are available to assess response of patients with anxiety disorders to neuromodulation therapies, including:

Anorexia nervosa is a disorder characterized by a pathological fear of food, eating and gaining weight, and sustained (more than three years) duration of significant underweight (body mass index [BMI] <18.5 kg/m²) [46,47]. Onset is usually peri-pubertal and anorexia nervosa mainly affects girls and women. The lifetime prevalence of anorexia nervosa in women is 2% to 4%, median duration is 5 to 7 years, and illness duration is more than 15 years in 30% [47]. The disorder has a high mortality rate, and only 10% to 30% of adults with anorexia nervosa recover through the best available therapies [46].

Patients with anorexia nervosa have limited treatment options, especially with prior treatment nonresponse [47]. Pharmacotherapy is largely ineffective and poorly accepted [46]. Patients underweight for three to five years have worse treatment response and outcome, possibly from the neurotoxic effects of starvation and stress hormones (cortisol) to the brain. This makes the first years from onset critical for successful intervention [47].

Anorexia nervosa is associated with structural changes such as reduced grey matter in fronto-limbic-striatal areas; functional changes with over-active limbic drives from the insula and amygdala, and diminished prefrontal activity; and hypoactivity of PFC regions during response inhibition and set-shifting tasks, a marker of poor inhibitory control that promotes anorexia nervosa chronicity. Improved, neuroscience-based treatments are needed that target the neural substrates of anorexia nervosa [46].

The Eating Disorder Examination Questionnaire (EDE-Q) measures eating disorder symptoms and general psychopathology. This has been the most common measure of response to therapy in the literature.

Bulimia nervosa is an eating disorder with recurrent episodes of binge eating characterized by eating large amounts of food in a discrete amount of time (i.e., within a two-hour period) and feeling of lack of control over eating during episodes; and recurrent inappropriate compensatory behavior to prevent weight gain (purging). The binge eating and compensatory behaviors both occur at least once a week, on average, for three months, and patient self-evaluation is unduly influenced by body shape and weight [28].

Similar to anorexia, reduced PFC activity may contribute to symptoms in bulimic patients through impaired inhibitory control of binge eating and purging; poor cognitive flexibility that perpetuates compulsive body checking and exercise; and obsessive pre-occupation with eating, weight and shape [46].

The lifetime prevalence of binge-eating disorder is roughly 5% of the U.S. adult population, and 1.2% to 4.5% are diagnosed with subthreshold binge-eating disorder and any binge eating. Compared to obese individuals who do not binge-eat, patients with binge-eating disorder are more prone to depression, anxiety, body dissatisfaction, low self-esteem, and social withdrawal [48].

Cognitive-behavioral therapy can reduce binge symptoms, but many patients lack response and treatment-related weight-loss is minimal. Effective and lasting treatments for binge-eating disorder are needed to alleviate symptoms of this psychological disorder and the medical consequences of obesity [48].

The safety guidelines for rTMS were determined in study participants who were largely free of antidepressant medications. While it is possible that psychotropic medication can affect the motor threshold, the National Network of Depression Centers and the American Psychiatric Association indicate there are no known absolute contraindications to psychotropic medication usage during rTMS. All medication use and change should be documented.

(https://pmc.ncbi.nlm.nih.gov/articles/PMC5846193 Last Accessed: July 24, 2025)

Level of Evidence: Expert Opinion/Consensus Statement

Potentially problematic medications commonly used in psychiatry and pain includes tricyclic antidepressants (nortriptyline, amitriptyline), antipsychotic drugs (chlorpromazine, clozapine), and antiviral medications. Abuse of alcohol or some illicit substances can increase cortical excitability and risk of a TMS-induced seizure. Withdrawal from CNS sedatives, including alcohol and benzodiazepines, increases seizure risk and patients must be asked about recent and current use. Recent substance abuse should be a contraindication to rTMS therapy [13]. The antidepressant bupropion is associated with increased seizure risk in some populations, raising concerns of its use during rTMS therapy. A systematic review found no evidence that supports excluding patients taking bupropion from rTMS [74].

Seizure induction is the most serious adverse effects with rTMS, but fewer than 25 cases have been reported worldwide to date. The estimated incidence of spontaneous seizures is 0.01% to 0.1% with rTMS, 0.1% to 0.6% with antidepressant drugs, and 0.07% to 0.09% in the general population. HF-rTMS is contraindicated in patients with seizure history. Safety of LF-rTMS has been demonstrated in patients with epilepsy but is not established in patients with depression and seizures. Seizure history is usually considered an absolute contraindication [8].

According to the National Network of Depression Centers and the American Psychiatric Association, a comprehensive review of the patient's health status (including historical and current medical, surgical, neurologic, and psychiatric conditions and medications) and physical examination are evaluation components to determine the medical safety and necessity of rTMS.

(https://pmc.ncbi.nlm.nih.gov/articles/PMC5846193 Last Accessed: July 24, 2025)

Level of Evidence: Expert Opinion/Consensus Statement

Absolute contraindications to rTMS are ferromagnetic metal in the head (e.g., plates or pins, bullets, shrapnel) and metallic hardware anywhere in the head, except the mouth (e.g., Cochlear implants, brain stimulators or electrodes, aneurysm clip) [8,13]. Relative contraindications include [8,13]:

Using conventional tDCS protocols (sessions ≤40 min, ≤2 times/day, ≤4 mA, electrodes that minimize skin burns), repeated sessions in more than 1,000 subjects receiving more than 33,200 sessions has not produced any reports of serious adverse events or irreversible injury to date. This includes diverse subject populations and patients from vulnerable populations [16,80]. Reports in the published evidence include patients receiving more than 100 tDCS sessions without adverse events from cumulative exposure; lack of unexpected or serious adverse events in more than 40 studies with more than 600 older adults; lack of decrements in behavior or mood in stroke populations; and no finding of drug agents that increase risk of serious adverse effects [80].

Seizure concerns with TMS have spread to tDCS, but evidence of increased seizure risk with tDCS is absent. tDCS produces static and not pulsed electric fields that are two orders of magnitude below those of rTMS, ECT and transcranial electrical stimulation, and thus, no apparent seizure risk [80].

Relapse is frequent after successful rTMS without maintenance treatment. An uncontrolled study reported a median 120 days until relapse, with 25%, 40%, 57%, and 77% relapsing at 2, 3, 4, and 6 months, respectively [87]. In contrast, rTMS maintenance as needed over 12 months in 257 patients was able to sustain remission in 71% of remitters, and response in 63% of responders [88]. Another study found 38% of rTMS responders relapsed within 24 weeks (mean: 109 days post-treatment). With reinstatement of as-needed rTMS, 73% met response and 60% met remission at 24 weeks [89].

A 20-week gradual taper of maintenance rTMS (from 3 sessions/week to 1 session/month) was compared to no maintenance, and relapse rates were 38% with maintenance versus 82% without maintenance. Five "clustered" maintenance sessions over three days every month extended the mean relapse to 10.8 months [8,90,91].

The available evidence suggests ECT is generally more rapid and effective in antidepression effects, with greater side effects and lower patient acceptance than rTMS. An open-label study compared efficacy and patient acceptance of ECT versus rTMS in outpatients with treatment-resistant depression. While both therapies significantly improved depression scores, benefits were greater with ECT than rTMS, and anxiety symptoms decreased with ECT but not rTMS. Patients who received ECT reported they would have chosen rTMS if it had been financially available, even when informed rTMS was less effective than ECT for treatment-resistant depression. The authors stated this reflects the level of stigma that surrounds ECT [29].

A meta-analysis compared the efficacy of ECT and rTMS in MDD. ECT was superior to HF-rTMS in response (64.4% vs 48.7%) and remission (52.9% vs 33.6%), and discontinuation was similar (8.3% vs 9.4%). ECT was superior in psychotic depression, but HF-rTMS and ECT were comparable in non-psychotic depression. ECT had a slight advantage over rTMS in overall improvement in HAM-D scores. Data on medium or long-term efficacy was insufficient. The same results were found with ECT versus LF-rTMS. ECT led to greater impairment in cognitive domains such as visual memory and verbal fluency. ECT and rTMS seemed comparably effective in patients with MDD without psychosis, but ECT was more effective with psychotic depression [92].

Other meta-analyses found larger differences favoring ECT for all outcomes, especially in MDD with psychotic features. rTMS response is poor in patients with ECT failure, and rTMS should be considered before pursuing ECT because patients lacking ECT response are unlikely to benefit from rTMS [8].

rTMS has mostly been evaluated as an add-on therapy to stable antidepressant regimens, and no evidence shows outcomes are improved by discontinuing antidepressants before rTMS. However, some evidence suggests that initiating antidepressants during rTMS promotes higher response and remission rates than rTMS alone [8,83].

In patients with treatment-resistant depression receiving LF-rTMS, responders showed lower psychomotor retardation at baseline than non-responders, and ex-smokers showed substantially higher response rates than current smokers [93].

The most frequent adverse effects were scalp pain during stimulation (40%) and transient headache after stimulation (30%). Both diminish with repeated treatment, respond to over-the-counter analgesics, and resulted in few study dropouts. The cognitive safety profile of rTMS appears benign; studies that assessed rTMS and cognitive performance found no differences between active and sham rTMS [8,94].

Although published evidence suggests that daily prefrontal rTMS has a significantly greater antidepressant effect than sham, with a magnitude of effect at least as large as antidepressant drugs, some issues remain unresolved [19]. The optimal hemisphere and position of coil placement remains unclear in depression treatment. Whether coil placement using individualized location via neuronavigated methods is superior to general algorithms for probabilistic positioning is not clear. Also, duration of rTMS treatment and patient follow-up are inadequate in many older studies.

Personality traits such as extraversion and neuroticism have attracted increasing attention as characteristics in predicting MDD response to pharmacotherapy, psychotherapy, and more recently, neuromodulation. Deep rTMS allows stimulation of deeper cortical regions, and the influence of personality dimensions in antidepressant response to deep rTMS in treatment-resistant depression was evaluated. After four weeks of daily deep rTMS (20 Hz, 3,000 pulses/session) of the left dorsolateral FPC, clinical remission was associated with higher baseline levels of agreeableness and conscientiousness. Levels of agreeableness and extraversion were linearly associated with antidepressant response. Neuroticism was not associated with antidepressant effects. Five-factor personality assessment may have prognostic value in deep rTMS for treatment-resistant depression, as agreeableness, extraversion, and conscientiousness were associated with reductions in depressive symptoms during treatment [95].

A variation of rTMS is theta-burst stimulation. In MDD, theta-burst stimulation is delivered at lower intensities than rTMS (70% to 80% of active motor threshold) and requires only one to three minutes of stimulation [8]. Studies of both theta-burst stimulation types suggest lasting effects on cortical excitability that exceed standard rTMS. Increased cortical excitability with intermittent theta-burst stimulation has persisted longer than HF-rTMS. With continuous theta-burst stimulation, persistence in reduction of cortical excitability has been less consistent. Enthusiasm over the promising rapid effects on synaptic plasticity with theta-burst stimulation should be tempered by the highly variable patient response [19].

Small randomized controlled trials of dorsolateral PFC theta-burst stimulation showed superiority over sham with left intermittent theta-burst stimulation but not right continuous theta-burst stimulation, with bilateral stimulation (left intermittent/right continuous) showing mixed results. Data on bilateral DMPFC theta-burst stimulation suggest similar outcomes between intermittent theta-burst stimulation and longer conventional 10-Hz rTMS. Randomized controlled trial of conventional rTMS and theta-burst stimulation are in progress [8].

A randomized sham-controlled trial of 20 sessions found intermittent theta-burst stimulation safe, and resulted in immediate statistically significant decreases in depressive symptoms versus sham. After the two-week double-blind protocol, only 28% of patients showed ≥50% reduction in depression scores, but response rates increased to 38% an extra two weeks later to indicate delayed clinical effects. Importantly, 30% of the responders were in clinical remission. The findings suggest that only four days of accelerated intermittent theta-burst stimulation treatment in treatment-resistant depression may lead to meaningful clinical responses within two weeks post-stimulation [96].

Intermittent theta-burst stimulation was examined in a crossover randomized sham-controlled trial of 50 suicidal, antidepressant-free patients with treatment-resistant depression. Patients received 20 intermittent theta-burst stimulation sessions over four days, and the alternate condition (active or sham) at study day 7. This study found a significant decrease in suicide risk, lasting up to one month, unrelated to active or sham stimulation and unrelated to depression response. A high placebo response to sham was noted in those receiving sham intermittent theta-burst stimulation before they crossed over to active intermittent theta-burst stimulation. None of the patients committed suicide until six months after intermittent theta-burst stimulation treatment [97].

A randomized controlled trial of magnetic seizure therapy versus unilateral ECT found similar rates of response (60% vs 40%) and remission (30% vs 40%). A large uncontrolled magnetic seizure therapy trial reported a 69% response rate and a 46% remission rate, similar to published ECT outcomes [98,99]. Studies evaluating magnetic seizure therapy versus sham stimulation, relapse following magnetic seizure therapy, or relapse prevention await publication [8].

Magnetic seizure therapy has shown lower rates of headaches and muscle aches than ECT, without noticeable anterograde or retrograde amnesia. Reorientation time (time from seizure and anesthesia emergence to fully oriented by person, place, and time) was briefer with magnetic seizure therapy (2 to 7 minutes) than ECT (7 to 26 minutes). The sole randomized controlled trial in magnetic seizure therapy (versus unilateral ECT) found comparable neuropsychological testing results after 12 treatments [98,100].

A review of six randomized sham-controlled trials found active tDCS significantly superior to sham in response (34% vs 19%), remission (23.1% vs 12.7%), and improvement in depression. Treatment-resistant depression and higher tDCS "doses" were negative and positive predictors of tDCS efficacy, respectively. The authors concluded the tDCS treatment effect size was comparable to those of rTMS and antidepressant drug treatment in primary care [104].

A meta-analysis of tDCS in MDD evaluated improvement in HAM-D scores, and a moderate-large effect size was found that favored active tDCS over sham. Two additional randomized sham-controlled trials were added, effect sizes were calculated for response and remission, and active tDCS was found not superior to sham in response or remission. Closer examination of the involved studies indicated that tDCS showed negative results in studies with high treatment resistance samples, and positive results in studies where subjects were lower in treatment non-response. Concurrent use of antidepressants, mood stabilizers, or benzodiazepines hampered patient response to tDCS. As such, tDCS seems efficacious for the treatment of depression, but it is not recommended in treatment-resistant depression or as an add-on to medication [75].

Several studies have evaluated persistence of acute effects and relapse following initial tDCS treatment, and strategies to prevent relapse. In one study, three months after completion of acute tDCS treatment, antidepressant effects persisted in almost half (47.8%) of patients without maintenance tDCS, and the need to prolong acute tDCS efficacy was noted [105].

Following clinical response to acute treatment, subjects received continuation tDCS every week for three months, and then every two weeks the final three months. Cumulative relapse-free survival was 83.7% at three months and 51.1% at six months; with decreased continuation tDCS at three months, relapse increased from 16.3% to 48.9%. Medication resistance was the greatest predictor of relapse during continuation tDCS [106].

Another study of tDCS relapse followed subjects 24 weeks. After response to acute treatment, subjects received up to nine tDCS sessions every two weeks for three months, followed by one tDCS session/month for three months. The mean response duration was 11.7 weeks, and relapse-free survival rate at 24 weeks was 47%. Patients with treatment-resistant MDD showed substantially worse 24-week survival versus patients with non-depression (10% vs 77%) [107].

In a double-blind trial, patients with MDD were randomized to active tDCS, sham tDCS and fluoxetine 20 mg. More rapid improvement in depressive symptoms occurred with active tDCS than with fluoxetine, and after six weeks, active tDCS and fluoxetine showed the same improvement in depression scores and were significantly superior to sham [108].

Another study randomized 120 patients with MDD to active or placebo sertraline 50 mg plus active or sham tDCS in a four-arm trial. The combined active sertraline plus active tDCS was superior to all other groups, both groups with one active agent (sertraline or tDCS) did not differ and were superior to the group with placebo plus sham. Of note is that sertraline and tDCS were started concurrently in non-resistant patients, which could account for the additive effect rather than blocked response seen with tDCS as an add-on to stable but marginally ineffective antidepressants [109].

At present, tDCS trials are reviewed and conducted in accordance with the Non-Significant Risk U.S. Food and Drug Administration (FDA) designation for medical devices [80]. tDCS is generally well-tolerated. Frequent adverse events (>50%) include reddening of the skin, itching, burning, heat, and tingling sensations at the site of electrode placement. Sporadically reported adverse events with similar rates between active tDCS and sham include nausea, euphoria, reduced concentration, and anxiety. Study dropout rates from adverse events are comparable for tDCS and sham (both 3%), but long-term data on safety and tolerability are lacking [8]. No post-acute side effects emerged during follow-up in the longer-term maintenance studies [105].

tDCS is recommended as a third-line treatment for MDD, with more research needed to establish optimal stimulation parameters [8]

A randomized sham-controlled trial evaluated 30-session (over six weeks) LF-rTMS in 25 patients with GAD. Patients undergoing active rTMS (versus sham) experienced higher response rates (71% vs 25%) and significant reductions in anxiety, worry, and depressive symptoms. At three months post-rTMS, 43% of active rTMS patients showed remission, versus 8% with sham. Response rates with active rTMS were maintained during follow-up, with some additional gains in remission rates. At post-treatment, right dorsolateral PFC activation was increased for active rTMS only, and changes in neuroactivation significantly correlated with changes in worry symptoms. These findings provide preliminary evidence that rTMS may improve GAD symptoms by modifying neural activity in the stimulation site [114].

A secondary analysis study was performed to determine if dorsolateral PFC neuromodulation improved emotion regulation in patients with GAD [114]. Statistically significant improvements in self-reported emotion regulation difficulties were found at post-treatment and three-month follow-up with active rTMS only. Improvements primarily involved the domains of goal-directed behaviors and impulse control, and were significantly associated with global clinician ratings of improvement. These preliminary results support rTMS as a treatment for GAD and suggest improved emotion regulation as a possible mechanism of change [34].

Bilateral LF-rTMS was evaluated in 13 patients with comorbid GAD and MDD, who received 24 to 36 rTMS sessions over five to six weeks. Following the last treatment, 11 of 13 (84.6%) patients achieved remission in anxiety symptoms (<5 on the GAD-7), and 10 of 13 patients (76.9%) achieved remission in depressive symptoms (<8 on the HAM-D). In this small pilot study, most patients with comorbid GAD/MDD achieved significant improvement in anxiety and depressive symptoms after bilateral rTMS [115].

In two randomized sham-controlled trials of LF-rTMS, patients with panic disorder were treated for two or four weeks as augmentation therapy. Active rTMS was superior to sham in reducing panic symptoms in one trial, and all patients showed improvement with no differences between active and sham in the other trial. The available data were insufficient to draw conclusions about rTMS efficacy in panic disorder [14].

The anxiolytic effects of deep rTMS in MDD patients were reviewed. Data from six open-label studies found that relative to baseline, large anxiolytic and antidepressant outcomes were obtained after 20 daily sessions of high-frequency deep rTMS. Unlike the antidepressant effect, anxiolytic effects were more heterogeneous across studies, and unrelated to concurrent antidepressant treatment [116].

Following three sessions per week for four weeks of deep rTMS or sham to the medial prefrontal cortex (mPFC) in 30 patients with PTSD, significant improvement in intrusive symptoms was observed (mean reduction: 31%) with active but not sham deep rTMS. This suggested the potential of deep rTMS to treat core PTSD symptoms and suggests a role of mPFC hypoactivity in presentation of the disorder [40,120].

The rate of current tobacco use disorder in clinical samples of patients with schizophrenia is 62%, significantly higher than in the general population and in other psychiatric disorders [138]. This high prevalence is thought to arise from nicotine-induced improvements in working memory, selective attention, and cognitive impairments in patients with schizophrenia. These nicotine effects are linked to stimulation of the alpha7-nicotinic acetlycholine receptor, which enhances thalamo-cortical functional connectivity and dopamine release regulation. Nicotine can modify neuroplasticity, and in contrast to nonsmokers, single-session, excitability-diminishing cathodal tDCS does not induce plasticity in patients with schizophrenia who smoke [139].

Neural plasticity induced by tDCS may differ between patients with schizophrenia who do and do not smoke. This was explicitly investigated in the first-ever study that evaluated the effects of cigarette smoking on tDCS response in patients with schizophrenia. Patients with schizophrenia with treatment-resistant auditory verbal hallucination received 10 sessions of frontotemporal tDCS (2 mA). Auditory verbal hallucinations symptoms decreased 20% in the entire sample, 46% in nonsmokers, and 6% in smokers. When response was defined as ≥25% symptom reduction, 83% of non-smokers vs 20% of smokers were responders. tDCS had no effect on cigarette use. This was the first published report to describe the impairing effect on tDCS response and efficacy in patients with hallucinatory schizophrenia who are actively smoking. Adjustment for age, medication use, and illness duration did not impact this effect [140].

Many (40% to 70%) patients with treatment-resistant schizophrenia remain symptomatic despite an adequate clozapine trial [141]. Brain stimulation approaches are promising for these patients, but clozapine co-administration has raised safety concerns. Studies have been reviewed to determine if this safety issue was resolved. While ECT is effective in clozapine-refractory schizophrenia, safety with the combination is not known. Until sufficient data accumulate, vigilance for adverse effects related to lowered seizure threshold, cognitive impairment, and cardiovascular events is required. In patients receiving clozapine, HF-rTMS over the dorsolateral PFC and LF-rTMS over the temporo-parietal cortex are safe, but efficacy is uncertain [141].

Relapse is common within days after smoking cessation, and combining the anti-craving effects of rTMS with nicotine replacement therapy (NRT) may attenuate withdrawal symptoms and craving to increase abstinence rates in smokers with severe nicotine dependence attempting to quit [143].

The day after quitting, smokers who failed to quit with standard treatments started NRT (21-mg patch) and received active 1-Hz rTMS or sham of the right dorsolateral PFC (10 sessions) for two weeks. Abstinence rates and craving were measured during combined treatment and up to 12 weeks after quitting. At the end of combined treatment, more participants were abstinent with active rTMS (89%) than sham (47%). Compulsive craving was significantly decreased with active rTMS but not sham, but no lasting effect was found [143].

Similar results were shown by an earlier randomized sham-controlled trial of 10-Hz rTMS to the left dorsolateral PFC for 10 daily sessions. Active rTMS, but not sham, led to significantly reduced cigarette consumption and blocked craving during treatment, but the effects dissipating after the rTMS sessions ended [144].

Deep rTMS was evaluated in 115 smokers randomized to 13 daily sessions of HF, LF, or sham deep rTMS to the lateral PFC and bilateral insula. HF (but not LF or sham) deep rTMS significantly reduced cigarette consumption and nicotine dependence. The combination of high-frequency deep rTMS and exposure to smoking cues enhanced cigarette reduction, with abstinence rates of 44% at the end of the treatment and 33% at six-month follow-up [145].

Results from two studies provided data to encourage larger trials in determining the clinical usefulness of this intervention. A single-session randomized sham-controlled trial of 1-mA anodal stimulation over the left dorsolateral PFC found that active tDCS (versus sham) significantly increased latency to smoking and decreased total number of cigarettes smoked [146].

Cathodal tDCS stimulation is a way to manipulate cortical excitability through inhibition. tDCS inhibition of the frontal-parietal-temporal association area (FPT) was examined for effects on attention bias to smoking-related cues and smoking behavior in a small short-term trial. Bilateral cathodal tDCS stimulation of the FPT area modestly decreased attention to smoking-related cues and significantly decreased daily cigarette use [147].

Most rTMS studies in alcohol use disorder have investigated craving but not drinking reduction. As with studies in nicotine dependence, most used HF-rTMS to target the dorsolateral PFC. The effect of rTMS on craving for alcohol was mixed, with some studies showing an effect while others do not [12].

More promising results came with using H-coil rTMS, but these were preliminary open trials. A small study of bilateral H coil stimulation of the PFC in patients with comorbid alcohol use disorder and dysthymic disorder found improvements in depressive symptoms and reductions in craving for alcohol. Another pilot study using H-coil stimulation of the medial PFC reported decreased alcohol intake by mean drinks per day and mean drinks on days of maximum alcohol use [12].

Four sessions of HF-rTMS was evaluated in a randomized sham-controlled trial of 17 residents of an alcohol detoxification program. At one-month follow-up, active (but not sham) HF-rTMS led to significantly decreased depressive symptoms and fast EEG frequencies, significantly increased inhibitory control and executive function, but no effect was found on craving or drinking reduction [148].

Published randomized controlled trials of tDCS in alcohol use disorder have shown mixed results. In 13 subjects with alcohol use disorder, one session of tDCS to the bilateral dorsolateral PFC significantly decreased alcohol craving (versus sham), which remained suppressed during exposure to alcohol cues [149].

In a relapse reduction study, 33 outpatients with severe alcohol use disorder received twice-daily 2 mA tDCS to the bilateral dorsolateral PFC for five days. At six-month follow-up, tDCS did not diminish craving, but 50% treated with active tDCS versus 11.8% with sham remained alcohol-abstinent. Active tDCS also led to improved perception of quality of life [150].

A disconnect between alcohol craving and alcohol use was also reported in a study of 13 patients with alcohol use disorder who received five weekly sessions of anodal tDCS to the left dorsolateral PFC. Active tDCS suppressed cravings but showed an unexpected trend for more frequent relapse; neither was found with sham [151].

Very promising results came from a pilot study in which 18 subjects with cocaine use disorder received 12 sessions of 10-Hz H-coil rTMS or sham to the bilateral PFC. After four weeks of stimulation, cocaine use in active and sham groups both decreased without difference, but decreases in cocaine use showed a significant difference at three months and six months post-baseline favoring the active rTMS group. Transient mild headaches were reported during stimulation [152].

Patients with addiction to crack cocaine received five sessions of active or sham bilateral dorsolateral PFC stimulation. The active tDCS group showed a higher percentage of abstinence at three-month follow-up versus sham [153]. This study protocol was replicated in a larger group of patients whose crack cocaine cravings were suppressed for at least one week by active versus sham tDCS. Effects on crack cocaine use were not studied [154].

Abstinent methamphetamine users received single-session anodal tDCS or sham to the right dorsolateral PFC. Active tDCS acutely reduced craving at rest, but significantly increased methamphetamine craving during methamphetamine-related cue exposure. Neither effect was found with sham [78].

The effect of rTMS on craving in substance addiction was evaluated in a review of eight randomized sham-controlled trials. Active TMS was significantly superior to sham, but only in studies targeting the right dorsolateral PFC with HF-rTMS. This finding was consistent throughout the reviewed studies [155].

The dorsolateral PFC is implicated in emotion regulation. HF-rTMS may improve food restriction and other maladaptive emotion regulation strategies in anorexia nervosa by remediating prefrontal region hypoactivity associated with poor impulse control and impaired cognitive flexibility in anorexia nervosa [46].

In a uncontrolled trial, 10 patients with anorexia nervosa received single-session HF-rTMS to the left dorsolateral PFC, and showed short-term improvements in anxiety, feeling full, and feeling fat [156]. In patients with chronic anorexia nervosa receiving HF-rTMS to the left dorsolateral PFC, significant improvements in eating disorder symptoms and mood were found after six months in 60%, with 40% deemed "recovered" based on psychometric scores. However, most participants lost weight, and therapeutic effects on psychopathology waned by 12 months [157].

Several single-session rTMS studies in small numbers of patients with bulimia were performed, and reported positive short-term effects of reduced binge-eating episodes and food craving [46]. Sham-controlled single-session studies of HF-rTMS of the left dorsolateral PFC reported inconsistent effects, with positive effects reported in one study of 38 patients and no effects in a study of 10 patients [158,159]. Randomized controlled trial with sham control using 20 rTMS sessions did not find benefit beyond sham [46,160].

A trial of 30 subjects with binge-eating disorder evaluated singles-session 2-mA tDCS or sham to the dorsolateral PFC. Compared with sham, tDCS decreased craving for sweets and savory proteins with greatest reductions in men; decreased total and preferred food intake by 11% and 17.5% regardless of sex; and reduced desire to binge eat in men on the day following active tDCS. Reductions in craving and food intake were predicted by eating less often for reward motives, and greater intent to restrict calories, respectively. The findings of enhanced cognitive control and/or decreased need for reward may reflect tDCS action on functional mechanisms, and should be further researched [48].

VNS "dose" is measured by electrical output, in mA of current. A comparison of low (0.25 mA), medium (0.5–1.0 mA) and high (1.25–1.5 mA) dose VNS found greater improvement in depressive symptoms with higher doses, greater sustained antidepressant response and less frequent suicide attempts with medium- and high-dose versus low-dose VNS [167].

Open-label studies found an average 31.8% response rate, but a randomized sham-controlled trial found no difference between VNS and sham at 12 weeks [168]. VNS efficacy may increase over time, and six trials comparing VNS to treatment as usual found at 12, 24, 48, and 96 weeks that VNS/treatment as usual remission rates were 3%, 5%, 10%, and 14% versus 1%, 1%, 2%, and 4% for treatment as usual alone [169]. Median time to VNS response was nine months in one study. In patients with response by 3 months (35%), response was maintained by 61.5% at 12 months, and 50% at 24 months [170,171].

High retention and low dropout rates suggest that patients without measurable response or remission gain therapeutic benefits undetected by depression rating scales [165]. Higher success rates with VNS in treatment-resistant depression were reported with careful patient selection to screen out comorbid personality disorders or active substance use disorders [26].

The most common adverse events after one year of VNS were voice alteration (69.3%), dyspnea (30.1%), pain (28.4%), and increased cough (26.4%), mostly from active VNS and resolved by halting stimulation. VNS tolerability improves over time, and adverse effect reports diminish. Compared to treatment as usual, patients with treatment-resistant depression receiving VNS showed lower all-cause mortality rates, including suicide [8].

Deep brain stimulation treatment for treatment-resistant depression remains investigational. Deep brain stimulation is mostly used to augment antidepressants, and few patients are free of psychotropic drugs when implanted [8]. The main deep brain stimulation brain targets are [8]:

The most common target in treatment-resistant depression is the subcallosal cingulate. Patients in deep brain stimulation trials are refractory to antidepressants, psychotherapy, and (often) ECT [8].

Short-term outcomes with deep brain stimulation show 30% to 60% response rates and 20% to 40% remission rates at three or six months, and medial forebrain bundle deep brain stimulation in seven patients reported response and remission rates of 85.7% and 57.1% [172,173].

As with VNS, two randomized sham-controlled trials of deep brain stimulation were halted early from lack of benefit, later seen as premature with recognition that efficacy increases with longer stimulation [8]. Outcomes of four studies showed depression severity was significantly reduced by 12 months. At 3, 6, and 12 months, response rates were 36.6%, 53.9%, and 39.9%, with remission rates 16.7%, 24.1%, and 26.3% [174].

Subcallosal cingulate deep brain stimulation studies of greater than one year duration reported response rates of 36% and 92% at one and two years, and a 58% remission rate at two years. A longer trial reported 62.5%, 46.2%, and 75% response rates at one, two, and three years, with 20% and 40% remission rates at two and three years. Long-term subcallosal cingulate and medial forebrain bundle deep brain stimulation led to improved health-related quality of life [173,175,176].

Maintaining Response

A small trial suggested that maintaining remission requires ongoing deep brain stimulation. Patients were treated to remission with subcallosal cingulate deep brain stimulation, and then randomized to on/off or off/on stimulation for three months. Most patients relapsed with deep brain stimulation turned off, and none with deep brain stimulation on [177].

Adverse Effects

Adverse events during long-term deep brain stimulation have been secondary to implantation (e.g., intracranial hemorrhage), perioperative risks (e.g., wound infection), effects of specific brain region stimulation, or deep brain stimulation parameters. Deep brain stimulation is generally well tolerated despite the risks of invasive neurosurgical procedures, and the pooled one-year dropout rate with subcallosal cingulate deep brain stimulation was 10.8%. Worsening neuropsychological performance has not been observed with any brain target, but improved cognitive performance has been reported [174].

Transient psychosis and hypomania have emerged when changing parameters during nucleus accumbens stimulation, resolved by switching parameters. No hypomania episodes were reported with subcallosal cingulate deep brain stimulation, including patients with bipolar disorder. Oculomotor adverse events of blurred vision and strabismus occurred in all patients receiving higher amplitude medial forebrain bundle deep brain stimulation. Reports of suicidality and completed suicide were deemed not device-related, but risk of suicidality may be increased by a history of pre-deep brain stimulation suicidality or major life stressors [8].

Deep brain stimulation is an emerging treatment option for patients with severe refractory Tourette syndrome. In 156 published cases, deep brain stimulation led to an overall 52.68% improvement of symptoms. Outcomes of controlled trials significantly favored stimulation versus sham. Significant symptom reductions were found across deep brain stimulation targets, with no differences in outcome indicating modulation of a common network. Despite the small patient numbers, deep brain stimulation for Tourette syndrome is a valid option for medically intractable patients [178].

Deep brain stimulation efficacy in severe, refractory OCD was reviewed in 25 trials. Outcome was change in Y-BOCS scores; response was ≥35% improvement in Y-BOCS (Table 3) [41]. Reported adverse effects included serious events during surgery (i.e., two seizures, three intracerebral hemorrhages), device-related adverse events, and stimulation-related events. Device-related events included breaks in stimulation leads and battery failure resulting in acute psychiatric symptoms until replacement [41]. Stimulation-related adverse events were acute increases in anxiety and hypomania, induced by changing stimulation parameters or battery depletion. All cases resolved with parameter adjustment. Cognitive problems resolved with time or parameter adjustment. No personality changes during deep brain stimulation were observed.

Neural circuits involved in OCD and depression include the VC/VS and anterior cingulate gyrus. VC/VS deep brain stimulation combined with anterior cingulotomy was evaluated in patients with refractory OCD and MDD. Mean baseline Y-BOCS scores were 34.7, decreasing to 23.0 at three months, and stable two years post-surgery at 19.0. These outcomes were comparable, but not superior, to published outcomes for VC/VS deep brain stimulation alone, and deep brain stimulation may be sufficient to control refractory OCD [179].

A few small deep brain stimulation studies have been performed in severely addicted patients. Five severely alcohol dependent patients received nucleus accumbens deep brain stimulation. Followed an average 38 months, all five showed significant craving reductions, two remained abstinent for more than four years, and three had truncated relapses [180]. Two patients with heroin dependence received nucleus accumbens deep brain stimulation; depressive and anxiety symptoms improved, with drug use reduced but not ceased [181].

Neuropathic pain is caused by abnormality, trauma, or disease of the somatosensory nervous system, resulting from various etiologies including traumatic or surgical injuries to peripheral nerves, infectious diseases (e.g., herpes zoster, postherpetic neuralgia), metabolic disorders (e.g., diabetic peripheral neuropathy), and injuries or diseases that affect the CNS such as stroke or spinal cord injury. Nearly 25% of people with chronic diabetes have neuropathic pain [186,187].

Among adults in the general population, an estimated 7% to 8% have chronic pain with neuropathic features, and 5% have moderate-to-severe pain with neuropathic features [188]. Neuropathic pain is associated with significant loss of productive time, withdrawal from the workforce, development of mood disorders, and disruption of family and social life [189]. Neuropathic pain is reported to be more severe than non-neuropathic pain and the management of chronic neuropathic pain is challenging; more than 50% of patients experience partial or no pain relief, and the adverse effects of medications used to manage the pain may limit their clinical utility in general and in elderly patients in particular [186,187].

In the United States, spinal cord injury occurs in 17,000 persons annually, and an estimated 282,000 patients are living with spinal cord injury. Six months after discharge, 27% of patients with spinal cord injury report pain that is severe enough to interfere with most daily activities. Spinal cord injury pain can develop at or below the level of spinal injury and does not correlate well with the magnitude or location of the lesion. Injury from gunshot is associated with more severe pain. Neuropathic, musculoskeletal, and/or visceral pain can contribute to spinal cord injury pain, and central neuropathic pain can develop weeks to months following injury [190,191].

Pain is reported by 11% to 55% of stroke survivors, and can arise from muscles, joints or viscera, or from the peripheral or central nervous system. The most common types of poststroke pain include hemiplegic shoulder pain, pain due to spasticity, and central poststroke pain. Along with spinal cord injury pain, central poststroke pain is a central neuropathic pain condition with pain
arising from a cerebrovascular lesion in the central somatosensory nervous system. Sensory descriptors used in patients with central poststroke pain include burning, aching, pricking, lacerating, shooting, squeezing, throbbing, sharp, stabbing, painful pins and needles, dull, and cramping. Treatment of central poststroke pain is difficult, due to the limited efficacy and dose-limiting side effects of available drugs [192].

In the United States, more than 300,000 spinal surgeries are performed annually, mainly in the lumbar spine, and as many as 100,000 result in failure where the patient experiences new-onset pain in addition to unresolved pain from the original problem. The level of pain is widely variable and may occur with neurologic deficits. Contributors to pain and the clinical features of failed back surgery syndrome include recurrent disk herniation, epidural abscess, scar tissue formation around the nerve root, facet joint syndrome, and muscle spasm. Patients with persistent radicular pain, usually from chronic nerve injury, greatly benefit from treatment that addresses the neuropathic pain [193].

Lumbosacral radiculopathy results from nerve root irritation and compression, resulting in a symptom distribution of the affected lumbar or sacral nerve root such as numbness, weakness, or paresthesia. Sciatica is the most common symptom of lumbar radiculopathy and refers to pain that radiates down the leg below the knee in the distribution of the sciatic nerve to indicate nerve root compromise from mechanical pressure or inflammation [194].

Chronic low back and neck pain are highly prevalent but largely unaddressed in brain stimulation trials because origin is usually multifactorial, psychosocial contribution is often prominent and the torso has less cortical representation [13].

Temporomandibular disorders (TMDs) involve persistent pain in the cheek and jaw area of the face and have an estimated 11% prevalence in community samples. TMD pain considerably impedes quality of life, with nearly 80% of patients with TMD reporting regular discomfort eating and more than 40% reporting difficulty performing their jobs. The importance of CNS factors in TMD etiology is established, and neuroimaging has revealed structural, functional, and neurochemical aberrations in TMD pain [195].

Phantom limb pain is painful or unpleasant sensation in the distribution of the lost or deafferentated body part. Phantom limb pain can resemble neuropathic pain with descriptors such as sharp, shooting, or electrical-like; or resemble nociceptive pain with dull, squeezing, or cramping pain. The pain can be widespread throughout the missing limb or confined to a smaller limb area. The prevalence of phantom limb pain following amputation may be as high as 85%. Phantom limb pain remains an extremely difficult pain condition to treat, and most pharmacologic therapies are not successful [196].

Postherpetic neuralgia is persistent neuropathic pain after the healing of herpes zoster infection. Risk of developing postherpetic neuralgia increases with age, and the duration can be months to years. Pain levels range from mild to excruciating, with pain severity placing the patient at risk for suicide. In the United States, postherpetic neuralgia is the most common cause of suicide in patients with chronic pain older than 70 years of age. In postherpetic neuralgia, peripheral and CNS sensitization generate pain characterized by severe or excruciating pain from light touch (allodynia) or severe spontaneous pain without allodynia. Postherpetic neuralgia pain is highly resistant to conventional pain treatment [197].

Complex regional pain syndrome is thought to result from injury to autonomic, central, and peripheral nervous systems. It presents as pain affecting one or multiple limbs without following a dermatomal distribution. The pain is disproportionate to the inciting injury, and pain severity can be disabling. The affected limb(s) undergo marked changes in skin color, texture, and/or temperature. Absence of nerve injury is termed CRPS-I, with CRPS-II describing complex regional pain syndrome that develops after nerve damage. However, recent studies have shown nerve damage in patients classified as CRPS-I, suggesting a false dichotomy [198]. The pathophysiology of complex regional pain syndrome is multifactorial, and while multimodal conventional treatment is indicated, some patients with complex regional pain syndrome do not respond to standard approaches and develop refractory disease [199].

Most patients with chronic refractory headaches have medication overuse headache, alleviated by detoxification from their headache medication. A subset remains with chronic refractory headache pain. Neuromodulation options can be considered in selected cases [200].

Migraine has a lifetime prevalence of 18%, and women experience higher past-year rates (13% to 18%) than men (5% to 10%). The main clinical feature of migraine is a unilateral or bilateral throbbing or pulsating headache of moderate-to-severe pain, often preceded by an aura. Other symptoms include gastrointestinal disturbances (nausea or vomiting) and intense sensitivity to light or sound. Medical treatment is particularly challenging, with non-response to high-dose preventive medications or side effects that complicate treatment course in many cases [200].

The main clinical feature of cluster headaches is severe, excruciating pain with unilateral distribution around and above the eye and along the side of the head or face; pain sensation described as sharp, boring, burning, throbbing, or tightening; and restlessness or agitation during the headache. Autonomic features associated with cluster headache include ipsilateral conjunctival injection and lacrimation, rhinorrhea or nasal blockage, and ptosis. The pain is more severe than any other headache type and ranks among the most severe known to humans, which has earned cluster headache the term "suicide headache" from reports of suicidal behavior in patients desperate to stop the pain [201]. Treatments for cluster headache are effective in some patients, but patients may not respond to medication, and ablative or destructive methods can be ineffective [200].

Fibromyalgia afflicts an estimated 5 million adults in the United States, of whom 80% to 90% are women [202]. The cardinal features of fibromyalgia are widespread pain and tenderness in multiple regions of the body, not attributable to another condition. Abnormal reactivity to painful stimuli or discomfort is characteristic of fibromyalgia. Pain with fibromyalgia is described as a persistent, diffuse, deep, aching, throbbing sensation in muscles that is most often continuous. Patients with fibromyalgia suffer from a complex symptom spectrum that includes pain, as well as sleep problems, fatigue, cognition difficulties, and depression [203].

The pathophysiology of fibromyalgia involves CNS dysfunction and neurotransmitter dysregulation that results in central sensitization to pain, where disordered processing in central afferent neurons accounts for the dominant symptoms of pain and tenderness in fibromyalgia [204].

According to a European consortium of experts, HF-rTMS of M1 contralateral to pain side displays definite analgesic efficacy for neuropathic pain; LF-rTMS is probably ineffective.

(https://www.sciencedirect.com/science/article/pii/S1388245719312799 Last Accessed: July 24, 2025)

Level of Evidence: A (Definitely effective or ineffective based on at least two Class I studies or one Class I study and at least two Class II studies)

Of nine randomized controlled trials evaluated, three had major limitations from using less than 1,000 pulses per session and studies used 5-day or 10-day stimulations. HF-rTMS (5–20 Hz) to M1 showed efficacy at short-term (≤1 week from last session) and mid-term (1 to 6 weeks from last session) follow-up. Pain scores were reduced by 20% to 45% following active stimulation, and 35% to 60% of patients were responders (>30% pain relief). These analgesic effects were obtained in neuropathic pain regardless of anatomical origin or whether the pain had central or peripheral nervous system involvement [27].

HF-rTMS to M1 contralateral to the pain side received a Level A recommendation for consistent analgesia in patients with neuropathic pain. This was based on results of 511 patients indicating significant efficacy, with pain relief >30% in 46% to 62% of patients, and >50% pain relief in 29% of patients [129]. There were also modest but significant analgesic long-term effects with repeated sessions.

In contrast, LF-rTMS to M1 contralateral to the pain side received a Level B recommendation (probably ineffective) in patients with neuropathic pain. This was based on results of 138 patients in six randomized sham-controlled trials, which consistently reported the absence of any significant analgesic effect of active rTMS versus sham [129].

Of the few rTMS studies targeting the dorsolateral PFC in neuropathic pain, the results were inconsistent and tended to show lack of efficacy [27].

A European consortium of experts has found probable efficacy of HF-rTMS of the left M1 or the left DLPFC in improving quality of life of patients with fibromyalgia.

(https://www.sciencedirect.com/science/article/pii/S1388245719312799 Last Accessed: July 24, 2025)

Level of Evidence: B (Probably effective based on at least two Class II studies or the combination of one Class I or II study and at least two Class III studies)

Neuropathic pain syndromes tend to show greatest benefit from rTMS to the M1, but some nonneuropathic chronic pain syndromes, such as CRPS-I or fibromyalgia, may have a neuropathic component. Focal lesions with defined onset, such as pain from shingles or trauma, have advantages of known localization and time of onset, but early cases often improve spontaneously which complicates treatment outcomes in research; thus, established cases with pain duration of at least one year are preferable [13].

rTMS should be applied contralaterally for localized neuropathic pain, or to the left hemisphere for widespread neuropathic pain or fibromyalgia. HF-rTMS (≥5 Hz) stimulation should be used, delivered by a figure-8 coil oriented parallel to the midline over M1 for ≥1 week with ≥1,000 pulses per session. Increasing the total pulses per session and repeating the sessions over several days or weeks may enhance rTMS analgesia [27].

One randomized controlled trial using 10 daily sessions of 10 Hz rTMS in patients with refractory CRPS-I pain found significantly greater analgesic effects with active than sham rTMS over the three-week treatment period [27,205].

HF-rTMS of M1 in patients with CRPS-I showed significant and rapid reductions in pain intensity, but persistence after stimulation was, on average, short-term. There was high variation in duration of treatment response, with total pain relief for three months post-rTMS reported [129].

HF-rTMS to the left M1 was evaluated in several controlled trials. Significant reductions in global pain on a numerical scale, and improvements in quality of life up to one month following 10 daily sessions were found, as well as extension of these effects for several months with maintenance sessions after completion of the original treatment course [129].

HF-rTMS to M1 was effective in five randomized sham-controlled trials at short term but not at mid-term follow-up. Results of one trial were negative for pain intensity but positive for improvement in quality of life, and another study with 14 sessions of left M1 rTMS over 21 weeks reduced pain for only one month beyond stimulation [27].

HF-rTMS to the left dorsolateral PFC also showed analgesic efficacy, with a mean 29% difference in pain relief between active and sham conditions. This degree of analgesia was also found with LF-rTMS to the left dorsolateral PFC. LF-rTMS to the right dorsolateral PFC showed mixed results, with open-label trials showing positive effect that was not replicated by a sham-controlled trial [129].

Most studies conducted in fibromyalgia using LF-rTMS (1 Hz) to the right dorsolateral PFC or HF-rTMS (10–20 Hz) to the left dorsolateral PFC reported results suggesting that analgesic effects were independent of antidepressant effects. Outcomes with highest-quality evidence show durable and clinically relevant pain reduction preceding improvements in depression [27].

HF-rTMS of the motor cortex, in contrast to the left dorsolateral PFC, has shown consistent efficacy in multiple trials that showed significant improvement in headache frequency, pain score and functional disability after active rTMS versus sham. LF-rTMS to M1 showed no benefit beyond placebo effect [129].

HF-rTMS of the left dorsolateral PFC showed mixed results in sham-controlled trials, with significant decreases in migraine attack frequency and intensity, and reductions in oral medication found up to 1 month after 12 sessions in one study, and 23 sessions of active rTMS over 8 weeks less effective than sham in decreasing the number of headache days in another trial [129].

HF-rTMS to the M1 in 19 sessions over six months led to significant relief of refractory facial pain, including cluster headache, with a 40% response rate at six-month follow-up. The analgesic effect remained when session duration was shortened from 20 to 10 minutes while keeping pulses per session (2,000) constant [206].

The effectiveness of maintenance rTMS to prolong analgesia and increase long-term pain reduction was evaluated in 55 patients with chronic refractory facial pain, including cluster headache, trigeminal neuropathic pain, and atypical facial pain. Subjects received 10 HF-rTMS treatments of M1 over two weeks, two sessions in week 3, one session in weeks 4 and 6, and monthly sessions the next 5 months. Pain intensity was measured on a 0–10 scale. Significant decreases from baseline to day 15 were found in permanent pain (5.2 to 3.2) and paroxysmal pain (8.6 to 4.5) intensities, and daily number of pain attacks (5.6 to 2.3). At day 15, 73% had a pain reduction ≥30%, decreasing to 40% at day 180. The analgesic effect was similar for all pain types, but significantly lower when session duration was shortened [206].

Land mine victims with phantom limb pain received 10-Hz rTMS or sham of the M1 contralateral to the amputated leg in 10 daily sessions. Follow-up, in days, began after the last session. Active rTMS led to significantly greater reduction in pain intensity scores at 15 days vs sham, but the difference was non-significant at 30 days. At 15 days, 70.3% with rTMS versus 40.7% with sham attained clinically significant pain reduction (≥30%). Traumatic amputees with phantom limb pain were able to experience clinically significant pain reduction following 10-session HF-rTMS without side effects, and further work will identify protocols to extend analgesia [207].

A randomized controlled trial evaluated high-frequency rTMS in 40 patients with postherpetic neuralgia. After 10 sessions of active or sham rTMS of the M1, active rTMS led to an overall 16.89% greater pain reduction than sham at final stimulation, and one and three months post-stimulation. Analgesic effects were associated with long-term improvement in quality of life [208].

HF-rTMS of the left dorsolateral FPC led to immediate, short-term analgesic effects and significant reduction of morphine consumption in small open-label trials [129].

Potentially problematic prescription medications used by some pain patients include tricyclics (e.g., nortriptyline, amitriptyline), antiviral medications, and antipsychotic medications (e.g., chlorpromazine, clozapine). Consuming or discontinuing commonly abused substances can increase cortical excitability and risk of a TMS-induced seizure. Withdrawal from sedatives (e.g., alcohol, benzodiazepines) increases seizure risk, so patients should be asked about recent and current use, and recent substance abuse should be an exclusion criterion [13].

The efficacy of 2 mA tDCS over the M1 or dorsolateral PFC (10 sessions over 2 weeks) was evaluated in a randomized sham-controlled trial of patients with chronic medically refractory fibromyalgia. Both M1 and dorsolateral PFC stimulation (but not sham) resulted in significant decreases in pain intensity and a positive benefit on quality-of-life measures immediately following the sessions, but only the M1 stimulation produced longer-lasting pain relief thorough 60-day follow up. The analgesic effects of dorsolateral PFC stimulation dissipated by 30-day follow up [210].

Prefrontal areas are involved in the affective/cognitive processing of pain, and tDCS to the dorsolateral PFC may be more relevant to modulation of the emotional experience of pain than for the somatosensory aspect of pain or pain intensity [18].

An optimized protocol for fibromyalgia pain treatment with tDCS has been published. High-definition 2-mA tDCS of M1 led to clinically significant benefits of ≥50% pain reduction in 50% of patients, with both responders and nonresponders benefiting from a cumulative treatment effect of significant pain reduction and improved quality of life over time. An estimated 15 high-definition tDCS sessions (median) are needed for ≥50% pain reduction [209].

TMD has a high prevalence and in many patients, pain and masticatory dysfunction persist despite a range of treatments. Women with chronic myofascial TMD pain received five sessions of active or sham 2-mA high-definition tDCS to the M1. Compared with sham, active tDCS led to significantly higher number of subjects with ≥50% pain reduction at four weeks, pain-free mouth opening at one-week follow-up, and sectional pain area, intensity and their sum measures contralateral to M1 stimulation during the treatment week. There were no changes on emotional measures. high-definition tDCS M1 stimulation led to clinically meaningful improvement in sensory-discriminative pain and motor measures up to four weeks post-treatment in patients with chronic myofascial TMD pain [211].

Eight subjects with unilateral lower or upper limb amputation and chronic phantom limb pain received active or sham 1.5 mA anodal tDCS to M1 daily for five days. Active but not sham tDCS decreased background pain and frequency of phantom limb pain paroxysms through one-week post-tDCS. Also, on each treatment day, active tDCS patients reported immediate pain relief and increased ability to move their phantom limb. Immediate patient response to sham was variable, with increased or decreased pain from baseline [212].

Burn pain severity and anxiety are highly correlated in burn patients, and a randomized sham-controlled trial assessed the effect of single-session cathodal 1-mA tDCS to the sensory cortex on pain anxiety in 60 patients with severe burn. Following stimulation, pain anxiety scores were significantly reduced with tDCS compared to sham. Post-stimulation assessment was followed by burn dressing; afterwards, pain anxiety scores remained significantly lower with tDCS versus sham. Acute pain anxiety reduction in patients with burns may be attained with cathodal tDCS [213].

Interferon-alpha (IFN) is a standard treatment for chronic hepatitis C infection, and many patients develop painful symptoms during the three- to six-month treatment course. BDNF is a marker of inflammatory-mediated pain, and tDCS was evaluated for effects on pain and plasma BDNF in 28 patients with hepatitis C infection. Following five days of 2-mA anodal stimulation of M1, active tDCS (compared with sham) reduced pain scores with a mean 56% reduction, enhanced BDNF levels with a mean increase of 37.48%, reduced chronic pain scores and analgesic use. Multivariate regression identified a significant interaction between pain reduction and BDNF level increase with active tDCS only [214].

Patients with neuropathic pain due to lumbosacral radiculopathy received rTMS (10 Hz) and tDCS (anodal 2-mA) in a sham-controlled crossover trial. Active rTMS was superior to tDCS and sham in pain intensity reduction. tDCS was not superior to sham, but its analgesic effects correlated to rTMS, suggesting common mechanisms of action. Lowered cold pain thresholds with rTMS correlated to its analgesic efficacy, but had no effect on individual neuropathic symptoms. rTMS was more effective than tDCS and sham in patients with neuropathic pain due to lumbosacral radiculopathy, and may modulate sensory and affective dimensions of pain [215].

The treatment efficacy of rTMS and tDCS of fibromyalgia was evaluated by reviewing 16 randomized sham-controlled trials. Significant improvements in fibromyalgia domains were found [204]. Overall treatment effect sizes were large for pain, sleep disturbance, fatigue, and tender points. Effect sizes were medium for depression and general health/function. rTMS showed a significantly greater effect size than tDCS. Primary motor cortex (M1) stimulation produced a subtle but greater effect size in pain reduction versus dorsolateral PFC, and dorsolateral PFC stimulation may be better for depression improvement. Clinically meaningful improvements in anxiety or cognition, or dose-response effect in fibromyalgia pain reduction, were not found for either modality.

Several studies delivered stimulation for two to four weeks, which possibly led to underestimated efficacy because rTMS in treatment-resistant MDD usually requires four to six weeks of daily treatment for durable response/remission.

Adverse events were generally minor, most commonly skin discomfort at the stimulation site, headache, neck pain, and dizziness. Many studies found no significant difference between active and sham stimulation. Some temporary neurobehavioral adverse effects were observed, including insomnia, sleepiness, restlessness, and worsening of depressive symptoms. Detrimental cognition effects or seizures were not observed [204].

A review of 17 open-label trials reported mixed pain outcomes, but analgesic efficacy may have been underestimated by brief study durations. Motor cortex stimulation has shown delayed, fluctuating analgesic effects, and randomized sham-controlled trials have found substantial pain relief at one year with initial non-responders [27].

Some, but not all studies have strongly correlated preoperative M1 rTMS response with follow-up response to motor cortex stimulation. Other studies found improved motor cortex stimulation outcomes using (versus not using) preoperative rTMS. Positive preoperative rTMS response predicts better motor cortex stimulation outcomes and can improve patient selection [27]. In 20 patients with refractory neuropathic pain followed a mean 6.1 years after motor cortex stimulation, mean pain reduction at any time ranged from 28.9% to 37.2%, and immediate post-rTMS improvement predicted long-term pain reduction. Greatest long-term benefit occurred in physical pain and dependence (autonomy in daily activities); little changed were disability, anxiety, stress and depression [217].

Motor cortex stimulation is safe and effective in central and peripheral neuropathic pain (≥40% to 50% pain reduction in approximately 50% of patients), with best outcomes in central post-stroke pain and neuropathic facial pain [218]. Benefit is established in spinal cord injury pain. Multiple randomized sham-controlled trials with ≥12-month follow-up show sustained response rates in 60% of patients with central and peripheral neuropathic pain. Intracranial motor cortex stimulation is more effective than non-invasive stimulation, and partial rTMS responders should be considered for motor cortex stimulation [13,219,220].

Deep brain stimulation study reviews found long-term pain relief in more than 80% of patients with intractable failed back surgery syndrome, 58% with post-stroke pain, and higher rates in peripheral neuropathic pain (phantom limb pain, polyneuropathies) [13].

In 59 patients with long-term follow-up after deep brain stimulation of the PVG (53.8%), VPL/VPM (12.8%), or both (33.3%), ≥50% pain reduction was attained in patients with phantom limb pain (89%), brachial plexus injury (50%), post-stroke (69%) and spinal cord injury (57%) pain, and chronic headache (54%) [228]. Complications included implantable pulse generator changes (42%), lead revisions (18%), lead erosion requiring removal, and infection, with two treated by antibiotics and five requiring device removal [222].

In seven open-label studies in peripheral or central neuropathic pain, mean pain reduction approached 50%, although pain relief varied largely across studies. The best outcomes occurred with somatosensory thalamus deep brain stimulation in peripheral neuropathic pain [27]. The dorsal anterior cingulate cortex is a potential deep brain stimulation target in chronic neuropathic pain that warrants further study, given its central role in cognitive and affective processing [189].

The American Headache Society practice recommendation for unilateral hypothalamic deep brain stimulation was downgraded to Negative B, based on a study showing active deep brain stimulation no different from sham in pain reduction and serious adverse effects during deep brain stimulation treatment [201].